<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>ValidAnytime Observatory — forecast-error ledger</title>
    <link>https://validanytime.com/observatory</link>
    <atom:link href="https://validanytime.com/observatory/rss.xml" rel="self" type="application/rss+xml" />
    <description>Dated forecast-error regime changes across US grid demand forecasts — anytime-valid alarms with a stated false-alarm budget. No vibes.</description>
    <language>en-us</language>
    <lastBuildDate>Mon, 06 Jul 2026 22:08:54 GMT</lastBuildDate>
    <item>
      <title>United States (Lower 48): forecast-error regime change from 2026-07-06 (ongoing)</title>
      <link>https://validanytime.com/observatory/US48#US48-2026-07-06T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/US48#US48-2026-07-06T0200</guid>
      <pubDate>Mon, 06 Jul 2026 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.59 vs 9.08; +51,744 MW error at onset; budget ≈1 false page per grid-year). Ongoing at last publish. We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-07-03 — Late-June 2026 heat wave</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-07-03T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-07-03T2000</guid>
      <pubDate>Fri, 03 Jul 2026 20:00:00 GMT</pubDate>
      <category>Late-June 2026 heat wave</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.43 vs 9.08; −6,896 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-07-05 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>United States (Lower 48): forecast-error regime change from 2026-06-12</title>
      <link>https://validanytime.com/observatory/US48#US48-2026-06-12T1400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/US48#US48-2026-06-12T1400</guid>
      <pubDate>Fri, 12 Jun 2026 14:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.31 vs 9.08; +44,138 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-06-14 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2026-06-09</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2026-06-09T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2026-06-09T0000</guid>
      <pubDate>Tue, 09 Jun 2026 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.23 vs 9.08; +3,405 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-06-10 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-03-04</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-03-04T2100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-03-04T2100</guid>
      <pubDate>Wed, 04 Mar 2026 21:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.67 vs 9.08; +495 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-03-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-02-25</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-02-25T1700</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-02-25T1700</guid>
      <pubDate>Wed, 25 Feb 2026 17:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.40 vs 9.08; +354 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-02-26 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Duke Energy (Carolinas): forecast-error regime change from 2026-02-11</title>
      <link>https://validanytime.com/observatory/DUK#DUK-2026-02-11T0600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/DUK#DUK-2026-02-11T0600</guid>
      <pubDate>Wed, 11 Feb 2026 06:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.12 vs 9.08; +10,033 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-02-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-02-01 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-02-01T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-02-01T2200</guid>
      <pubDate>Sun, 01 Feb 2026 22:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.32 vs 9.08; −4,248 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-02-03 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-01-30 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-01-30T2300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-01-30T2300</guid>
      <pubDate>Fri, 30 Jan 2026 23:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.19 vs 9.08; −3,396 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-31 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2026-01-27 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2026-01-27T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2026-01-27T0300</guid>
      <pubDate>Tue, 27 Jan 2026 03:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.92 vs 9.08; −9,995 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-30 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Tennessee Valley Authority: forecast-error regime change from 2026-01-26 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/TVA#TVA-2026-01-26T1400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/TVA#TVA-2026-01-26T1400</guid>
      <pubDate>Mon, 26 Jan 2026 14:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.30 vs 9.08; −2,837 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-28 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2026-01-26 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2026-01-26T0600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2026-01-26T0600</guid>
      <pubDate>Mon, 26 Jan 2026 06:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.25 vs 9.08; −11,290 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-28 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2026-01-26 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2026-01-26T0400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2026-01-26T0400</guid>
      <pubDate>Mon, 26 Jan 2026 04:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.47 vs 9.08; −11,143 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-28 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Tennessee Valley Authority: forecast-error regime change from 2026-01-26 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/TVA#TVA-2026-01-26T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/TVA#TVA-2026-01-26T0000</guid>
      <pubDate>Mon, 26 Jan 2026 00:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.06 vs 9.08; −4,468 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-28 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Duke Energy (Carolinas): forecast-error regime change from 2026-01-25 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/DUK#DUK-2026-01-25T1200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/DUK#DUK-2026-01-25T1200</guid>
      <pubDate>Sun, 25 Jan 2026 12:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.07 vs 9.08; +4,492 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-26 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Duke Energy (Carolinas): forecast-error regime change from 2026-01-24 — January–February 2026 cold wave (Winter Storm Fern)</title>
      <link>https://validanytime.com/observatory/DUK#DUK-2026-01-24T2300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/DUK#DUK-2026-01-24T2300</guid>
      <pubDate>Sat, 24 Jan 2026 23:00:00 GMT</pubDate>
      <category>January–February 2026 cold wave (Winter Storm Fern)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.51 vs 9.08; +5,188 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-26 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2026-01-18</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2026-01-18T0600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2026-01-18T0600</guid>
      <pubDate>Sun, 18 Jan 2026 06:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.16 vs 9.08; −2,946 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Public Service Colorado (Xcel): forecast-error regime change from 2026-01-10</title>
      <link>https://validanytime.com/observatory/PSCO#PSCO-2026-01-10T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PSCO#PSCO-2026-01-10T2200</guid>
      <pubDate>Sat, 10 Jan 2026 22:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.47 vs 9.08; +1,158 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-12 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Public Service Colorado (Xcel): forecast-error regime change from 2026-01-10</title>
      <link>https://validanytime.com/observatory/PSCO#PSCO-2026-01-10T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PSCO#PSCO-2026-01-10T0000</guid>
      <pubDate>Sat, 10 Jan 2026 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.49 vs 9.08; +1,233 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-12 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Bonneville Power (Northwest): forecast-error regime change from 2026-01-05</title>
      <link>https://validanytime.com/observatory/BPAT#BPAT-2026-01-05T1800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/BPAT#BPAT-2026-01-05T1800</guid>
      <pubDate>Mon, 05 Jan 2026 18:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.96 vs 9.08; +371 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-08 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Los Angeles (LADWP): forecast-error regime change from 2026-01-05</title>
      <link>https://validanytime.com/observatory/LDWP#LDWP-2026-01-05T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/LDWP#LDWP-2026-01-05T0300</guid>
      <pubDate>Mon, 05 Jan 2026 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.26 vs 9.08; −2,540 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Los Angeles (LADWP): forecast-error regime change from 2026-01-03</title>
      <link>https://validanytime.com/observatory/LDWP#LDWP-2026-01-03T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/LDWP#LDWP-2026-01-03T0000</guid>
      <pubDate>Sat, 03 Jan 2026 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.92 vs 9.08; −2,306 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Los Angeles (LADWP): forecast-error regime change from 2026-01-02</title>
      <link>https://validanytime.com/observatory/LDWP#LDWP-2026-01-02T1600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/LDWP#LDWP-2026-01-02T1600</guid>
      <pubDate>Fri, 02 Jan 2026 16:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 13.43 vs 9.08; −2,250 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2026-01-01</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2026-01-01T2300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2026-01-01T2300</guid>
      <pubDate>Thu, 01 Jan 2026 23:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.18 vs 9.08; +6,442 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Bonneville Power (Northwest): forecast-error regime change from 2026-01-01</title>
      <link>https://validanytime.com/observatory/BPAT#BPAT-2026-01-01T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/BPAT#BPAT-2026-01-01T0200</guid>
      <pubDate>Thu, 01 Jan 2026 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.38 vs 9.08; +428 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2026-01-03 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2025-12-18</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2025-12-18T0900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2025-12-18T0900</guid>
      <pubDate>Thu, 18 Dec 2025 09:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.11 vs 9.08; −5,158 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-12-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2025-12-18</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2025-12-18T0700</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2025-12-18T0700</guid>
      <pubDate>Thu, 18 Dec 2025 07:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.61 vs 9.08; −4,458 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-12-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2025-12-11</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2025-12-11T1900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2025-12-11T1900</guid>
      <pubDate>Thu, 11 Dec 2025 19:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.73 vs 9.08; −4,849 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-12-12 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2025-12-02</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2025-12-02T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2025-12-02T0300</guid>
      <pubDate>Tue, 02 Dec 2025 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.47 vs 9.08; −8,369 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-12-03 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2025-09-13</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2025-09-13T1800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2025-09-13T1800</guid>
      <pubDate>Sat, 13 Sep 2025 18:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.59 vs 9.08; −2,736 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-09-15 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2025-08-11</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2025-08-11T2100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2025-08-11T2100</guid>
      <pubDate>Mon, 11 Aug 2025 21:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.36 vs 9.08; −915 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-08-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2025-08-05</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2025-08-05T1800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2025-08-05T1800</guid>
      <pubDate>Tue, 05 Aug 2025 18:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.33 vs 9.08; −1,681 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-08-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2025-07-19</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2025-07-19T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2025-07-19T0300</guid>
      <pubDate>Sat, 19 Jul 2025 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.21 vs 9.08; −1,280 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-07-20 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2025-07-07</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2025-07-07T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2025-07-07T0200</guid>
      <pubDate>Mon, 07 Jul 2025 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.53 vs 9.08; −689 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-07-09 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2025-06-25</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2025-06-25T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2025-06-25T2000</guid>
      <pubDate>Wed, 25 Jun 2025 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.69 vs 9.08; +1,984 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-06-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2025-06-22</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2025-06-22T1600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2025-06-22T1600</guid>
      <pubDate>Sun, 22 Jun 2025 16:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 17.05 vs 9.08; −1,829 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-06-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2025-06-18</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2025-06-18T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2025-06-18T0300</guid>
      <pubDate>Wed, 18 Jun 2025 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.31 vs 9.08; −11,130 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-06-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2025-06-11</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2025-06-11T0400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2025-06-11T0400</guid>
      <pubDate>Wed, 11 Jun 2025 04:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.20 vs 9.08; −667 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-06-12 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2025-02-25</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2025-02-25T1600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2025-02-25T1600</guid>
      <pubDate>Tue, 25 Feb 2025 16:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.11 vs 9.08; −4,372 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2025-02-22</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2025-02-22T1800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2025-02-22T1800</guid>
      <pubDate>Sat, 22 Feb 2025 18:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.35 vs 9.08; −4,456 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Tennessee Valley Authority: forecast-error regime change from 2025-02-21</title>
      <link>https://validanytime.com/observatory/TVA#TVA-2025-02-21T1300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/TVA#TVA-2025-02-21T1300</guid>
      <pubDate>Fri, 21 Feb 2025 13:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.10 vs 9.08; −2,387 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2025-02-20</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2025-02-20T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2025-02-20T2000</guid>
      <pubDate>Thu, 20 Feb 2025 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.59 vs 9.08; −4,266 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2025-02-20</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2025-02-20T1500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2025-02-20T1500</guid>
      <pubDate>Thu, 20 Feb 2025 15:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.99 vs 9.08; −3,538 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Arizona Public Service: forecast-error regime change from 2025-02-07</title>
      <link>https://validanytime.com/observatory/AZPS#AZPS-2025-02-07T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/AZPS#AZPS-2025-02-07T2000</guid>
      <pubDate>Fri, 07 Feb 2025 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.33 vs 9.08; +887 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-09 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southern Company (Southeast): forecast-error regime change from 2025-01-31</title>
      <link>https://validanytime.com/observatory/SOCO#SOCO-2025-01-31T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SOCO#SOCO-2025-01-31T0500</guid>
      <pubDate>Fri, 31 Jan 2025 05:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.22 vs 9.08; −2,113 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-02-01 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2025-01-23 — January 2025 cold wave</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2025-01-23T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2025-01-23T2000</guid>
      <pubDate>Thu, 23 Jan 2025 20:00:00 GMT</pubDate>
      <category>January 2025 cold wave</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.66 vs 9.08; −4,490 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-24 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Los Angeles (LADWP): forecast-error regime change from 2025-01-22 — January 2025 cold wave</title>
      <link>https://validanytime.com/observatory/LDWP#LDWP-2025-01-22T0800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/LDWP#LDWP-2025-01-22T0800</guid>
      <pubDate>Wed, 22 Jan 2025 08:00:00 GMT</pubDate>
      <category>January 2025 cold wave</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.41 vs 9.08; −1,407 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2025-01-21 — January 2025 cold wave</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2025-01-21T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2025-01-21T2200</guid>
      <pubDate>Tue, 21 Jan 2025 22:00:00 GMT</pubDate>
      <category>January 2025 cold wave</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.63 vs 9.08; −7,285 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-24 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2025-01-21 — January 2025 cold wave</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2025-01-21T1800</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2025-01-21T1800</guid>
      <pubDate>Tue, 21 Jan 2025 18:00:00 GMT</pubDate>
      <category>January 2025 cold wave</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.46 vs 9.08; +4,961 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Tennessee Valley Authority: forecast-error regime change from 2025-01-10</title>
      <link>https://validanytime.com/observatory/TVA#TVA-2025-01-10T1700</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/TVA#TVA-2025-01-10T1700</guid>
      <pubDate>Fri, 10 Jan 2025 17:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 13.27 vs 9.08; −2,055 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2025-01-06</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2025-01-06T0900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2025-01-06T0900</guid>
      <pubDate>Mon, 06 Jan 2025 09:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.21 vs 9.08; +3,577 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-07 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2025-01-03</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2025-01-03T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2025-01-03T0100</guid>
      <pubDate>Fri, 03 Jan 2025 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.22 vs 9.08; +1,606 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2025-01-04 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2024-12-22</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2024-12-22T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2024-12-22T0300</guid>
      <pubDate>Sun, 22 Dec 2024 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 12.17 vs 9.08; +1,473 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-12-24 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2024-12-21</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2024-12-21T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2024-12-21T0200</guid>
      <pubDate>Sat, 21 Dec 2024 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.96 vs 9.08; −3,710 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-12-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-12-09</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-12-09T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-12-09T0100</guid>
      <pubDate>Mon, 09 Dec 2024 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.44 vs 9.08; +1,511 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-12-10 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-11-29</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-11-29T1900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-11-29T1900</guid>
      <pubDate>Fri, 29 Nov 2024 19:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.17 vs 9.08; −1,552 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-11-30 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2024-07-14</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2024-07-14T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2024-07-14T0300</guid>
      <pubDate>Sun, 14 Jul 2024 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.10 vs 9.08; −938 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-07-15 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2024-07-09</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2024-07-09T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2024-07-09T0000</guid>
      <pubDate>Tue, 09 Jul 2024 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.82 vs 9.08; −9,589 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-07-10 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2024-07-08</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2024-07-08T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2024-07-08T0500</guid>
      <pubDate>Mon, 08 Jul 2024 05:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.98 vs 9.08; −889 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-07-10 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2024-06-29</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2024-06-29T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2024-06-29T0000</guid>
      <pubDate>Sat, 29 Jun 2024 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.62 vs 9.08; +19,639 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-06-30 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Los Angeles (LADWP): forecast-error regime change from 2024-06-27</title>
      <link>https://validanytime.com/observatory/LDWP#LDWP-2024-06-27T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/LDWP#LDWP-2024-06-27T2000</guid>
      <pubDate>Thu, 27 Jun 2024 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.70 vs 9.08; −3,072 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-06-30 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2024-06-22</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2024-06-22T1000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2024-06-22T1000</guid>
      <pubDate>Sat, 22 Jun 2024 10:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.01 vs 9.08; +1,276 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-06-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2024-06-18</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2024-06-18T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2024-06-18T0200</guid>
      <pubDate>Tue, 18 Jun 2024 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.76 vs 9.08; +5,929 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-06-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Florida Power &amp; Light: forecast-error regime change from 2024-03-12</title>
      <link>https://validanytime.com/observatory/FPL#FPL-2024-03-12T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/FPL#FPL-2024-03-12T0100</guid>
      <pubDate>Tue, 12 Mar 2024 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.21 vs 9.08; +1,073 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-03-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-03-03</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-03-03T1400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-03-03T1400</guid>
      <pubDate>Sun, 03 Mar 2024 14:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.33 vs 9.08; +3,326 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-03-07 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-02-17</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-02-17T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-02-17T0000</guid>
      <pubDate>Sat, 17 Feb 2024 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.54 vs 9.08; +3,114 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-02-18 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-02-15</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-02-15T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-02-15T0500</guid>
      <pubDate>Thu, 15 Feb 2024 05:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.16 vs 9.08; +3,381 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-02-16 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2024-01-21</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2024-01-21T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2024-01-21T0500</guid>
      <pubDate>Sun, 21 Jan 2024 05:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.68 vs 9.08; +1,249 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2024-01-20</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2024-01-20T0600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2024-01-20T0600</guid>
      <pubDate>Sat, 20 Jan 2024 06:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.38 vs 9.08; −610 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>PacifiCorp East (Utah/Wyoming): forecast-error regime change from 2024-01-17 — January 2024 Arctic outbreak</title>
      <link>https://validanytime.com/observatory/PACE#PACE-2024-01-17T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/PACE#PACE-2024-01-17T0000</guid>
      <pubDate>Wed, 17 Jan 2024 00:00:00 GMT</pubDate>
      <category>January 2024 Arctic outbreak</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.65 vs 9.08; −800 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-18 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>United States (Lower 48): forecast-error regime change from 2024-01-14 — January 2024 Arctic outbreak</title>
      <link>https://validanytime.com/observatory/US48#US48-2024-01-14T1000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/US48#US48-2024-01-14T1000</guid>
      <pubDate>Sun, 14 Jan 2024 10:00:00 GMT</pubDate>
      <category>January 2024 Arctic outbreak</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.34 vs 9.08; +21,962 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-17 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-01-14 — January 2024 Arctic outbreak</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-01-14T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-01-14T0500</guid>
      <pubDate>Sun, 14 Jan 2024 05:00:00 GMT</pubDate>
      <category>January 2024 Arctic outbreak</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.08 vs 9.08; +5,361 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-16 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-01-13 — January 2024 Arctic outbreak</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-01-13T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-01-13T0000</guid>
      <pubDate>Sat, 13 Jan 2024 00:00:00 GMT</pubDate>
      <category>January 2024 Arctic outbreak</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 15.19 vs 9.08; +2,693 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-16 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Bonneville Power (Northwest): forecast-error regime change from 2024-01-10</title>
      <link>https://validanytime.com/observatory/BPAT#BPAT-2024-01-10T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/BPAT#BPAT-2024-01-10T0000</guid>
      <pubDate>Wed, 10 Jan 2024 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.37 vs 9.08; +596 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-11 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2024-01-07</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2024-01-07T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2024-01-07T0100</guid>
      <pubDate>Sun, 07 Jan 2024 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.36 vs 9.08; +1,431 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-09 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2024-01-02</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2024-01-02T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2024-01-02T2200</guid>
      <pubDate>Tue, 02 Jan 2024 22:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.59 vs 9.08; +6,558 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2024-01-03 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2023-12-24</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2023-12-24T0900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2023-12-24T0900</guid>
      <pubDate>Sun, 24 Dec 2023 09:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.59 vs 9.08; +5,028 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2023-12-25 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2023-12-22</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2023-12-22T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2023-12-22T0100</guid>
      <pubDate>Fri, 22 Dec 2023 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 12.22 vs 9.08; −4,493 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2023-12-24 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2023-06-30</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2023-06-30T0100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2023-06-30T0100</guid>
      <pubDate>Fri, 30 Jun 2023 01:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.49 vs 9.08; −8,227 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2023-07-01 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2023-06-05</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2023-06-05T0400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2023-06-05T0400</guid>
      <pubDate>Mon, 05 Jun 2023 04:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.18 vs 9.08; −6,409 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2023-06-07 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2023-02-04</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2023-02-04T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2023-02-04T0200</guid>
      <pubDate>Sat, 04 Feb 2023 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.33 vs 9.08; −6,869 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2023-02-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2022-12-26 — December 2022 cold (Winter Storm Elliott)</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2022-12-26T1400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2022-12-26T1400</guid>
      <pubDate>Mon, 26 Dec 2022 14:00:00 GMT</pubDate>
      <category>December 2022 cold (Winter Storm Elliott)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.36 vs 9.08; −12,207 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-12-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2022-12-25 — December 2022 cold (Winter Storm Elliott)</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2022-12-25T2300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2022-12-25T2300</guid>
      <pubDate>Sun, 25 Dec 2022 23:00:00 GMT</pubDate>
      <category>December 2022 cold (Winter Storm Elliott)</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.39 vs 9.08; −5,966 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-12-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2022-08-30</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2022-08-30T0400</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2022-08-30T0400</guid>
      <pubDate>Tue, 30 Aug 2022 04:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.19 vs 9.08; −4,060 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-08-31 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2022-08-16</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2022-08-16T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2022-08-16T2200</guid>
      <pubDate>Tue, 16 Aug 2022 22:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.53 vs 9.08; −4,113 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-08-18 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2022-07-06</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2022-07-06T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2022-07-06T0200</guid>
      <pubDate>Wed, 06 Jul 2022 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.81 vs 9.08; −2,828 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-07-08 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2022-02-08</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2022-02-08T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2022-02-08T0200</guid>
      <pubDate>Tue, 08 Feb 2022 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.15 vs 9.08; −2,892 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-02-09 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2022-01-11</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2022-01-11T2200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2022-01-11T2200</guid>
      <pubDate>Tue, 11 Jan 2022 22:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.82 vs 9.08; −2,546 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-01-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2022-01-04</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2022-01-04T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2022-01-04T0200</guid>
      <pubDate>Tue, 04 Jan 2022 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.85 vs 9.08; +1,214 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2022-01-05 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2021-12-26</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2021-12-26T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2021-12-26T0500</guid>
      <pubDate>Sun, 26 Dec 2021 05:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.15 vs 9.08; −3,170 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-12-27 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2021-12-02</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2021-12-02T2300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2021-12-02T2300</guid>
      <pubDate>Thu, 02 Dec 2021 23:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.22 vs 9.08; −4,866 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-12-04 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2021-08-25</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2021-08-25T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2021-08-25T0000</guid>
      <pubDate>Wed, 25 Aug 2021 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.34 vs 9.08; −20,881 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-08-26 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2021-07-05</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2021-07-05T2100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2021-07-05T2100</guid>
      <pubDate>Mon, 05 Jul 2021 21:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.50 vs 9.08; −4,272 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-07-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2021-07-04</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2021-07-04T1600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2021-07-04T1600</guid>
      <pubDate>Sun, 04 Jul 2021 16:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.51 vs 9.08; +5,059 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-07-06 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2021-06-28</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2021-06-28T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2021-06-28T2000</guid>
      <pubDate>Mon, 28 Jun 2021 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.02 vs 9.08; +1,289 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-07-01 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2021-06-21</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2021-06-21T1900</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2021-06-21T1900</guid>
      <pubDate>Mon, 21 Jun 2021 19:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.41 vs 9.08; −6,791 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-06-22 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2021-06-06</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2021-06-06T2100</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2021-06-06T2100</guid>
      <pubDate>Sun, 06 Jun 2021 21:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.74 vs 9.08; +1,493 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-06-09 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2021-05-26</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2021-05-26T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2021-05-26T2000</guid>
      <pubDate>Wed, 26 May 2021 20:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.02 vs 9.08; +4,038 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-05-28 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2021-05-25</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2021-05-25T0000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2021-05-25T0000</guid>
      <pubDate>Tue, 25 May 2021 00:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.21 vs 9.08; +3,245 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-05-26 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>New York (NYISO): forecast-error regime change from 2021-05-24</title>
      <link>https://validanytime.com/observatory/NYIS#NYIS-2021-05-24T0300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/NYIS#NYIS-2021-05-24T0300</guid>
      <pubDate>Mon, 24 May 2021 03:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.08 vs 9.08; +1,308 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-05-25 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2021-05-22</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2021-05-22T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2021-05-22T0200</guid>
      <pubDate>Sat, 22 May 2021 02:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.23 vs 9.08; +3,768 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-05-23 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2021-02-17 — Winter Storm Uri</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2021-02-17T0200</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2021-02-17T0200</guid>
      <pubDate>Wed, 17 Feb 2021 02:00:00 GMT</pubDate>
      <category>Winter Storm Uri</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.02 vs 9.08; −4,494 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-02-19 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2021-02-15 — Winter Storm Uri</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2021-02-15T0600</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2021-02-15T0600</guid>
      <pubDate>Mon, 15 Feb 2021 06:00:00 GMT</pubDate>
      <category>Winter Storm Uri</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.98 vs 9.08; +4,096 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-02-18 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>United States (Lower 48): forecast-error regime change from 2021-02-12 — Winter Storm Uri</title>
      <link>https://validanytime.com/observatory/US48#US48-2021-02-12T0500</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/US48#US48-2021-02-12T0500</guid>
      <pubDate>Fri, 12 Feb 2021 05:00:00 GMT</pubDate>
      <category>Winter Storm Uri</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.37 vs 9.08; +25,127 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-02-14 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Texas (ERCOT): forecast-error regime change from 2021-02-10 — Winter Storm Uri</title>
      <link>https://validanytime.com/observatory/ERCO#ERCO-2021-02-10T2000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/ERCO#ERCO-2021-02-10T2000</guid>
      <pubDate>Wed, 10 Feb 2021 20:00:00 GMT</pubDate>
      <category>Winter Storm Uri</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 11.46 vs 9.08; +6,088 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-02-13 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>Southwest Power Pool: forecast-error regime change from 2021-01-19</title>
      <link>https://validanytime.com/observatory/SWPP#SWPP-2021-01-19T1000</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/SWPP#SWPP-2021-01-19T1000</guid>
      <pubDate>Tue, 19 Jan 2021 10:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 10.52 vs 9.08; −2,754 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-01-20 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
    <item>
      <title>MISO (Midwest): forecast-error regime change from 2021-01-14</title>
      <link>https://validanytime.com/observatory/MISO#MISO-2021-01-14T1300</link>
      <guid isPermaLink="true">https://validanytime.com/observatory/MISO#MISO-2021-01-14T1300</guid>
      <pubDate>Thu, 14 Jan 2021 13:00:00 GMT</pubDate>
      <category>Forecast-error regime change</category>
      <description>The day-ahead demand forecast&apos;s error crossed its alarm threshold (peak log-e 9.84 vs 9.08; −4,226 MW error at onset; budget ≈1 false page per grid-year). Back in-band for a full window as of 2021-01-16 (display heuristic, not a certificate of absence). We detect that the operator&apos;s own forecast changed regime, not that the grid failed.</description>
    </item>
  </channel>
</rss>